. Author manuscript; available in PMC: 2018 Nov 30.

Published in final edited form as: Stat Med. 2018 Jul 18;37(27):4007–4021. doi: 10.1002/sim.7891

TABLE 3.

Simulation results for estimation of covariate-adjusted log-odds ratio relating MCP-1 and spontaneous abortion (2500 trials, true value = 0.20)

	Mean Bias (Median Bias)	SD(IQR)	Mean SE	MSE	95% CI Coverage
PE only
Logistic regression ignoring MCP-1 errors	−0.097	0.047	0.047	0.012	0.447
LRF	0.014	0.099	0.099	0.010	0.958
LRA	0.013	0.099	0.098	0.011	0.958
DFA	0.005	0.092	0.094	0.009	0.959
ME only
Without replicates^a
Logistic regression ignoring MCP-1 errors	−0.025	0.064	0.062	0.005	0.919
LRF	0.027	0.108	0.104	0.012	0.970
LRA	0.027	0.107	0.104	0.012	0.970
DFA	0.003	0.084	0.095	0.007	0.970
With replicates
LRF	0.005	0.076	0.074	0.006	0.954
LRA	0.004	0.076	0.074	0.006	0.954
DFA	0.001	0.074	0.073	0.005	0.954
PE and ME
Without replicates
Logistic regression ignoring MCP-1 errors	(−0.106)	(0.061)	-	-	0.355
LRF	(0.062)	(0.306)	-	-	0.987^c
LRA	(0.062)	(0.302)	-	-	0.987^d
DFA^b	(0.053)	(0.290)	-	-	0.986
With replicates
LRF	0.014	0.103	0.103	0.011	0.962
LRA	0.013	0.102	0.102	0.011	0.962
DFA	0.005	0.095	0.098	0.009	0.964

Excludes 2 trials in which LRF produced extreme log-OR estimates (>2).

non-bias-adjusted version of estimator used because bias adjustment frequently flipped sign of log-OR estimate (74.8% of trials).

Excludes 70 trials in which variance-covariance matrix was not positive definite.

Excludes 79 trials in which variance-covariance matrix was not positive definite.

CI, confidence interval; DFA, discriminant function approach; IQR, interquartile range; LRA, logistic regression with approximate maximum likelihood; ME, measurement error; OR, odds ratio; PE, processing error; SD, standard deviation; SE, standard error; MSE, mean squared error.